Fruit Fly Optimization Algorithm Based on Random Numbers Mutation Operator

The fruit fly optimization algorithm has low convergence precision and easily falls into local optimum. Its moving step value is not easy to determine, which is weak to solve complex optimization problems. Therefore, an improved fruit fly optimization algorithm is proposed. The improved algorithm employs a random numbers mutation operator to disturb the best position coordinates value of each generation of the fruit fly population as the moving step, which is perturbed by setting the dynamic inertia disturbance factor. Thus, the value of the moving step length is adaptive. The experimental results of eight high-dimensional peak function show that the improved algorithm has higher convergence precise and faster convergence speed than those of the comparison algorithm, and the success rate of optimization is 100% under the higher target precision. Therefore, the improved algorithm which a random numbers mutation operator is employed can increase the discrete degree of the individual distribution of the fruit fly and expand the diversity of the fruit fly population so that the improved algorithm can improve the abilities of seeking the global excellent result and evolution speed.